Extracellular Vesicles Carrying miR-887-3p Promote Breast Cancer Cell Drug Resistance by Targeting BTBD7 and Activating the Notch1/Hes1 Signaling Pathway

Objective Chemoresistance remains the primary reason threatening the prognosis of breast cancer (BC) patients. Extracellular vesicles (EVs) contribute to chemoresistance by carrying microRNAs (miRNAs). This study investigated the mechanism of miR-887-3p mediated by EVs in BC cell drug resistance. Methods MDA-MB-231-derived EVs were extracted and identified. BC cells were treated with different concentrations of doxorubicin, cisplatin, and fulvestrant, and the cell survival was evaluated. PKH26-labeled EVs were cocultured with BC cells, and the uptake of EVs was observed. miR-887-3p expression in BC cells and EVs was detected. After silencing miR-887-3p in MDA-MB-231 cells, BC cells were treated with EV-inhi to observe drug resistance. The target gene of miR-887-3p was predicted and verified. The levels of downstream Notch1/Hes1 pathway were detected. Xenograft experiment was conducted to evaluate the effect of EVs on the growth and drug resistance in vivo. Results MDA-MB-231-derived EVs enhanced the drug resistance of BC cells. EVs carried miR-887-3p into BC cells. miR-887-3p expression was elevated in BC cells and EVs. miR-887-3p inhibition reduced the drug resistance of BC cells. miR-887-3p targeted BTBD7. Overexpression of BTBD7 partially reversed the drug resistance of BC cells caused by EV treatment. EV treatment increased the level of Notch1/Hes1, and overexpression of BTBD7 decreased the level of Notch1/Hes1. In vivo experiments further validated the results of in vitro experiments. Conclusion EVs carrying miR-887-3p could target BTBD7 and activate the Notch1/Hes1 signaling pathway, thereby promoting BC cell drug resistance. This study may offer novel insights into BC treatment.


Introduction
Breast cancer (BC) is the main cause of cancer-related mortality in women throughout the world, and the number of BC-caused deaths in 2017 worldwide was 60,728, occupying clinical outcome of BC [8]. Therefore, it is imperative to figure out the molecular mechanisms of BC onset and drug resistance.
Extracellular vesicles (EVs) are bioactive molecular shuttles packaged by proteins, lipids, and nucleic acids, which can modulate the tumor microenvironment (TME) by interacting with adjacent cells [9]. Tumor-derived EVs function as crucial mediators of intercellular communication between tumor cells and normal stromal cells in the local and distant TME, thereby facilitating tumor progression and drug resistance [10]. MicroRNAs (miRNAs) represent the most widely studied molecules in EVs [11]. miRNA has about 22 nucleotides in size, which functions as an antisense RNA to downregulate the expression of target genes at the posttranscriptional level [12]. Evidence has demonstrated that miRNAs circulate in body fluids in a highly stable and acellular form in cancer patients, which may be due to their incorporation in EVs, making them useful as new diagnostic and prognostic markers [13]. Intriguingly, aberrant miRNA expression has also been commonly accepted as a promising biomarker for the drug resistance of BC [14]. Drug-resistant BC cells can release EVs through specific miRNA cell-to-cell metastasis and spread resistance to sensitive cells [15]. Rolf Sokilde et al. have analyzed the BC miRNA microarray GSE131599 and found that 11 miRNAs including miR-887 are notably overexpressed in ER + BC [16]. Lv et al. have shown that miR-887 is highly expressed in BC cell lines, and inhibition of miR-887 enhances the sensitivity of BC patients to 5-Fu treatment [17]. However, whether EVs can affect the drug resistance of BC cells by carrying miR-887-3p is unclear yet. This study investigated the effect of miR-887-3p carried by EVs on the drug resistance of BC, which shall confer novel insights into the clinical management of BC.

Ethics Statement.
This study got the approval and supervision of the Ethics Committee of Anhui No. 2 Provincial People's Hospital. Animal experiments were approved by the institutional ethical guidelines. Significant efforts were made to minimize the number of animals and their pain.
2.5. EV Extraction. EVs were isolated and purified with polyethylene glycol-(PEG-) based enrichment and ultracentrifugation as reported previously [18]. A 2 × stock solution of PEG6000 (Sigma-Aldrich) was prepared. MDA-MB-231 cell medium was centrifuged at 500 × g for 5 min and then at 2000 × g for 30 min at 4°C to eliminate cellular debris and large apoptotic bodies. Next, the conditioned medium (CM) was filtered through a 0.22 μm filter (Merck Millipore, Billerica, MA, USA) to clear away microvesicles. Afterwards, an equal amount of prepared 2 × PEG solution was added to the CM and then incubated at 4°C for 12 h while vibrating. The next day, the samples were centrifuged at 3220 × g and 4°C for 1 h and resuspended with particle-free phosphatebuffered saline (PBS). Subsequently, EVs were ultracentrifuged at 110,000 × g and 4°C for 70 min with an Optima XPN-100 ultracentrifuge (Beckman Coulter, Chaska, MN, USA) and then washed to remove contaminated proteins and PEG. Finally, the EVs were resuspended again in particle-free PBS for immediate use or were stored at -80°C. The EVs were identified by a transmission electron microscope (TEM) and nanoparticle tracking analysis (NTA).
2.6. Uptake of EVs. MCF-7, HCC1937, and BT474 cells were stained using CellTrace CFSE cell proliferation kits (Invitrogen). EVs were labeled using a PKH26 Fluorescent Cell Linker kit (Sigma-Aldrich). Briefly, 1 mL PKH26 solution (1 : 1000) was mixed with EVs (20 μg protein) for 20 min, and the mixture was washed with PBS and centrifuged at 1000000 × g for 70 min. Then, the PKH26-labeled EVs were added into CFSE-labeled cells for 24 h incubation, and the uptake of EVs was observed under a confocal fluorescence microscope.
2.14. Animal Studies. A total of 48 BALB/c mice (aged 4-6 weeks; weighing 18 to 25 g) purchased from Guangdong Medical Laboratory Animal Center (Foshan, Guangdong, China) were kept under specific pathogen-free conditions. MCF-7 cells were resuspended with 50 μL PBS and added with 50 μL Matrigel at 5 × 10 6 cells/mL and subcutaneously injected into each mouse. After bearing the tumors for 7 days, 100 μg EVs extracted from MDA-MB-231 cells (BC +EV group), extracted from miR-887-3p inhibitortransfected MDA-MB-231 cells (BC+EV-inhi group), or extracted from NC-transfected MDA-MB-231 cells (BC +EV-NC group) were injected into each mouse via a tail vein. Six aliquot injections were administered at 2-day intervals. PBS was used as the vehicle (BC group). The dose of EVs was based on the previous literature [20]. Mice were injected with doxorubicin (10 mg/kg, i.v., push) once on days 12, 15, and 18 after subcutaneous implantation of EVs in MCF-7 cells. The growth of BC xenograft in mice was monitored every 5 days, and 20 days later, tumor growth was monitored every 3 days. At 32 days post-implantation, the mice were euthanized by carbon dioxide asphyxiation.   7 Disease Markers sections were visualized using 3,3-diaminobezidine (DA1010, Solarbio, Beijing, China). Five fields of view at 200 × magnification were randomly captured for each replicate using an inverted microscope (Nikon, Tokyo, Japan).

Statistical Analysis.
All experiments were performed three times. Data were exhibited in the mean ± standard deviation. Statistical analysis was performed with GraphPad Prism 8 software (GraphPad, San Diego, CA, USA). Data were analyzed using the one-way or two-way analysis of variance (ANOVA), followed by Tukey's post hoc test. The p value < 0.05 was regarded as statistically significant.

Extraction and Identification of EVs.
EVs were extracted from MDA-MB-231 cells by ultracentrifugation. The biomarker proteins of EVs were determined using western blot analysis, and the results demonstrated that CD63 and CD81 were only expressed in EVs; TSG101 and Alix were highly expressed in EVs than in cell lysates, while GM130 and Cytochrome C was not expressed in EVs (Figure 1(a)).
Under the TEM, it was observed that the extracted EVs were uniformly distributed and presented in vesicles of different sizes (Figure 1(b)). The size of the extracted EVs was identified by NTA, and EV particles were mainly distributed at about 100 nm with a concentration of 2:5 × 10 6 particles/ mL (Figure 1(c)). These results indicated that the MDA-MB-231-derived EVs were extracted successfully.

Inhibition of miR-887-3p in EVs Weakened Drug
Resistance and Growth of BC Cells Induced by EVs. To further verify the role of miR-887-3p carried by MDA-MB-231-derived EVs in drug resistance of BC cells, we added EV-inhi (15 μg/mL) into MCF-7, HCC1937, and BT474 cells for 24 h incubation. Then, EV-treated MCF-7, BT474, and HCC1937 cells were treated with 0.5 μM Dox, 2.5 μM Cis, and 0.5 μM Ful, respectively, followed by cell viability, colony formation ability, and cell apoptosis detection. The results showed that compared with EV-NC, inhibition of miR-887-3p expression in EVs significantly reduced the viability of MCF-7, BT474, and HCC1937 cells under drug treatment and increased   3.5. miR-887-3p-Targeted BTBD7. To determine the downstream mechanism of miR-887-3p regulating drug resistance in BC cells, we predicted target genes of miR-887-3p through RNAInter, Targetscan, and ENCORI ( Figure 5(a)). The Coexpedia database was used to search the coexpression relationship of genes to further screen target genes. According to the coexpression relationship score, we screened the three target genes MDM4, BTBD7, and MAP3K1 with the highest score (score = 4:677, 2.550, and 2.126) ( Figure 5(b)). BTBD7 expression is downregulated in human BC cell lines and tissues, and BTBD7 inhibits the proliferation and invasion/migration of BC cells [21]. Downregulation of BTBD7 can promote apoptosis and increase the sensitivity of NSCLC cells to paclitaxel [22]. Starbase predicted that there is a specific binding site between miR-887-3p and BTBD7 3 ′ UTR (all p < 0:05, Figure 5(c)). Based on this sequence, we designed a luciferase reporter plasmid based on CMV, which contained the binding sites of miR-887-3p mimic and miR-negative control with wt or mut BTBD7 3 ′ UTR, respectively. The results showed that miR-887-3p could bind to 3 ′ UTR of BTBD7 specifically (all p < 0:05, Figure 5(d)). The RNA pull-down assay further proved the binding relationship between miR-887-3p and BTBD7.

Disease Markers
Moreover, BTBD7 inhibits BC cell proliferation, invasion, migration, and tumor metastasis by inactivating the Notch1 pathway [21]. The Notch1 pathway can promote drug resistance of cancer cells including gastric cancer cells, ovarian cancer cells, and prostate cancer cells [23][24][25][26][27]. Therefore, we detected the levels of Notch1 pathwayrelated proteins Notch1 and Hes1 in BC cells, and the results showed that the levels of Notch1/Hes1 were increased after EV treatment, while they were decreased after BTBD7 overexpression (Figure 6(e)). Taken together, MDA-MB-231derived EVs carrying miR-887-3p targeted BTBD7 and activated the Notch1/Hes1 signaling pathway to promote BC cell drug resistance.   13 Disease Markers derived EVs in the growth of MCF-7 cells in vivo, we measured the growth and weight of transplanted tumors in nude mice. Compared with doxorubicin treatment alone (BC group), MDA-MB-231-derived EV treatment (BC+EV group) significantly increased tumor growth rate and tumor volume in mice, while miR-887-3p inhibition in EVs (BC +EV-inhi group) weakened the effects (p < 0:05, Figures 7(a) and 7(b)). The immunohistochemical results showed that after EV treatment, BTBD7-positive cells were decreased while Ki-67-positive cells were increased in MCF-7-transplanted tumors, which were reversed by inhibition of miR-887-3p (p < 0:05, Figure 7(c)). Western blot demonstrated that the levels of Notch1/Hes1 were activated after EV treatment, while inhibition of miR-887-3p reversed such effects (all p < 0:01, Figure 7(d)). Briefly, MDA-MB-231-derived EVs promoted BC cell drug resistance in vivo.

Discussion
Although the better treatment and earlier diagnosis of BC have improved in past years, resistance to chemotherapy or radiotherapy, recurrence, and distant metastasis still exist and lead to undesirable prognosis [28,29]. Strikingly, previous demonstration supported that miRNA-EV delivery decreased the sensitivity of BC cells to doxorubicin [30]. In light of this, we made a hypothesis prior to the experiments that there may be a miRNA-EV delivery between miR-887-3p and MDA-MB-231-derived EVs in BC cell drug resistance. Collectively, such a conclusion could be drawn that miR-887-3p incorporated in MDA-MB-231-derived EVs promoted BC cell drug resistance by targeting BTBD7 and activating the Notch1/Hes1 signaling pathway (Figure 8).
The first major finding in the present study was that MDA-MB-231-derived EVs enhanced BC cell drug resistance. EVs secreted by metastatic tumors promote tumor invasion, inhibit immune responses, and enhance angiogenesis and chemoresistance by delivering RNA between cells [31]. It has recently been described that anticancer drugs strongly increase tumor cell secretion of EVs, facilitating the chemoresistance and posttherapy relapse through signaling pathway activation and inflammation induction [32]. Recently, it reported the miRNA-EV delivery in BC increased EV-induced migration and drug resistance [33]. It was verified that EVs promote angiogenesis, support the growth and expansion of tumors, and transmit anticancer drugs outside the BC cells, leading to drug resistance [34]. Moreover, tumors grow and evolve through a constant crosstalk with the surrounding microenvironment, and